Abstract: System, methods and computer program products for identifying a featured product based on product information submitted by retailers and manufacturers are described. In some implementations, a retailer and a manufacturer can supply product information to an ad serving system. Based on the product information supplied by the retailer and the manufacturer, a product matching the product can be identified. The matching product can then be presented on a content of the retailer that links to a product page of the matching product.

Abstract: System, methods and computer program products for identifying a featured product based on product information submitted by retailers and manufacturers are described. In some implementations, a retailer and a manufacturer can supply product information to an ad serving system. Based on the product information supplied by the retailer and the manufacturer, a product matching the product can be identified. The matching product can then be presented on a content of the retailer that links to a product page of the matching product.

Abstract: Techniques for increasing the amount of information that may be displayed in a graph produced by a visualization system and simplifying the specification of the information. With regard to nodes of the graph, locations relative to the displayed node at which values of attributes of the displayed node are effective may be specified. 17 locations may be specified using values that indicate compass directions and whether the location is inside or outside the boundaries of the node. Among the attributes whose locations may be specified are content attributes whose values are strings or images and action attributes which specify actions that occur in response to mousing actions at the specified locations. Specification of attributes of nodes and also of links is simplified by the use of inheritance. Also disclosed are an API for setting and reading attributes and a technique for producing compressed representations of nodes at rendering time.

Abstract: Techniques employed in displays of graphs with large numbers of nodes for viewing details of an area of interest while preserving context. The techniques take advantage of parent-child and sibling relationships. Sibling nodes and children of sibling nodes are displayed in rows. The center of interest is at the center of a row and nodes at the ends of the rows overlap. A parent node may have a node scroller associated with it for viewing the parent's child nodes one page of nodes at a time. The node scroller includes paging buttons and may also include a scroll bar with a slider. The node scroller may be displayed on the parent node itself or on the links connecting the parent node to the page of children currently being displayed. Also disclosed are details of the implementation of a preferred embodiment of the node scroller and details of an application programmer's interface (API) for creating and manipulating node scrollers.